Potential conflict of interest: Nothing to report.
Wnt drives stem cell-mediated repair response after hepatic injury
Article first published online: 19 SEP 2013
© 2013 by the American Association for the Study of Liver Diseases
Volume 58, Issue 5, pages 1847–1850, November 2013
How to Cite
Nejak-Bowen, K. N. and Monga, S. P.S. (2013), Wnt drives stem cell-mediated repair response after hepatic injury. Hepatology, 58: 1847–1850. doi: 10.1002/hep.26579
- Issue published online: 30 OCT 2013
- Article first published online: 19 SEP 2013
- Accepted manuscript online: 20 JUN 2013 08:21AM EST
- Manuscript Accepted: 6 JUN 2013
- Manuscript Revised: 6 MAY 2013
- Manuscript Received: 13 MAR 2013
Huch M, Dorrell C, Boj SF, van Es JH, Li VS, van de Wetering M, et al. In vitro expansion of single Lgr5+ liver stem cells induced by Wnt-driven regeneration. Nature 2013;494:247-250. (Reprinted with permission.)
The Wnt target gene Lgr5 (leucine-rich-repeat-containing G-protein-coupled receptor 5) marks actively dividing stem cells in Wnt-driven, self-renewing tissues such as small intestine and colon, stomach and hair follicles. A three-dimensional culture system allows long-term clonal expansion of single Lgr5+ stem cells into transplantable organoids (budding cysts) that retain many characteristics of the original epithelial architecture. A crucial component of the culture medium is the Wnt agonist RSPO1, the recently discovered ligand of LGR5. Here we show that Lgr5-lacZ is not expressed in healthy adult liver, however, small Lgr5-LacZ+ cells appear near bile ducts upon damage, coinciding with robust activation of Wnt signalling. As shown by mouse lineage tracing using a new Lgr5-IRES-creERT2 knock-in allele, damage-induced Lgr5+ cells generate hepatocytes and bile ducts in vivo. Single Lgr5+ cells from damaged mouse liver can be clonally expanded as organoids in Rspo1-based culture medium over several months. Such clonal organoids can be induced to differentiate in vitro and to generate functional hepatocytes upon transplantation into Fah−/− mice. These findings indicate that previous observations concerning Lgr5+ stem cells in actively self-renewing tissues can also be extended to damage-induced stem cells in a tissue with a low rate of spontaneous proliferation.
Liver stem cells are thought to reside in biliary ducts, are analogous to hepatoblasts during hepatic development, and being bipotential can give rise to both hepatocytes and biliary epithelial cells. The molecular basis for the maintenance and differentiation of the liver stem cells remain unidentified. Wnt signaling has been shown to be important in hepatoblasts, atypical ductular reaction and in rat liver stem cells.[1-3] However, the exact identity of liver stem cells remains an enigma and necessitates recognition of specific and reliable markers along with a suitable in vitro model to characterize their role and regulation in hepatic health and disease. This was recently addressed by Huch et al., where they demonstrate the appearance and expansion of a periportal Lgr5+ cell population upon liver damage that undergoes in vitro and in vivo expansion and differentiation to relatively mature epithelial cells of the liver in a 3D culture system. Lgr5, a Wnt target gene that is restricted to a rare population of proliferating cells in adult tissues, has been implicated as a stem cell marker in multiple organs capable of self-renewal.
The authors use the Axin2-LacZ mice, which demonstrate basal Wnt/β-catenin activity in pericentral hepatocytes only. Upon carbon tetrachloride (CCl4)-induced pericentral liver injury, the authors detected smaller β-galactosidase-positive cells in the periportal region, which occurred discordantly from the pericentral Wnt gene expression, which was down-regulated. Some of the up-regulated genes after such injury included Wnt6, several Rspondins, and Lgr5. This prompted analysis of reporter activity in Lgr5-lacZ reporter mice exposed to CCl4, which showed consistent periportal expression of Lgr5 in small cells near ducts; these cells shared a similar gene expression profile with biliary epithelial cells, including up-regulation of multiple Wnt target genes. The authors also performed lineage-tracing experiments by breeding Lgr5-IRES-creERT2 mice with Rosa26-lacZ Cre reporter mice and activating Cre recombinase expression after liver injury with CCl4, MCDE, or 3,5-diethoxycarbonyl-1,4-dihydrocollidine (DDC). The authors discovered small LacZ+ cells that evolved into hepatocytes and bile ducts. However, the consistency of appearance and differentiation of Lgr5+ cells among models is lacking; for example, while β-galactosidase-positive cells after CCl4 were limited to a few small cells and then a few hepatocytes, after DDC the entire duct was strongly positive with only the occasional hepatocyte demonstrating any immunoreactivity to this marker. Lack of intermediate timepoints in the DDC model makes it hard to interpret whether Lgr5+ expression is turned on in every biliary epithelial cell after injury or an entire duct lining is being derived from an occasional Lgr5+ liver stem cell.
The authors next established a novel organoid culture where bile duct fragments were cultured in Matrigel along with factors such as HGF, EGF, FGF10, nicotinamide, and R-spondin1 (Rspo1), a ligand for Lgr5. These cultures formed cysts that evolved into larger hepatic organoids, which continued to express Lgr5 and biliary markers and could be maintained for more than 12 months with weekly passaging as long as the media contained EGF, Rspo1, and nicotinamide. Flow-sorted, single Lgr5-LacZ+ cells replicated the above results, demonstrating the stemness of these cells, and is a major highlight of the report. Lgr5+ organoids were found to have expression profiles resembling an adult liver, although they still expressed high levels of progenitors markers such as Sox9, Cd44, and Prom1 and mature hepatocyte markers were either absent or only weakly expressed. Thus, the authors conclude that the organoid culture by default is biased towards biliary differentiation, which is not surprising since bile ducts appear to harbor these cells in the first place. Hepatocyte maturation could be induced by inhibition of probiliary Notch and TGF-β signaling, which also led to a decrease in progenitor markers and an increase in mature hepatocyte marker expression along with glycogen accumulation, albumin secretion, low-density lipoprotein (LDL) uptake, and P450 function. Intriguingly, lingering CK19 expression indicated a persistent ductal phenotype. Thus, the Lgr5+ cells are truly bipotential in this cell population, although bias toward induction of a default biliary phenotype was observed (Fig. 1). It would have been more convincing if a direct comparison of stemness and differentiation of Lgr5+ cells to Sox9+/Lgr5- or CK19+/Lgr5- cells could be made in the organoid cultures, as it would underscore the heterogeneity of biliary epithelial cells in terms of their stem cell characteristics.
Finally, Huch et al. transplanted organoids derived from single Lgr5+ cells cultured in hepatocyte differentiation media for 9 days, into the fumarylacetoacetate hydrolase (Fah−/−) mutant mice. Fah+ nodules representing transplanted cell-derived colonies were found within the liver in only 5 of the 15 mice. The repopulation ranged anywhere between 0.1 to 1% of total hepatic parenchyma and led to only a partial rescue of the enzymatic defect in Fah−/− animals. This was drastically lower than engraftment and rescue of Fah−/− animals by transplantation of freshly isolated hepatocytes. However, the engrafted Lgr5+ derived hepatocytes increased recipient animal survival significantly and did not lead to any oncogenic events. Similarly, it was interesting to note that the in vivo hepatic milieu led to sufficient differentiation of organoids to hepatocytes, since no CK19 expression was detected in engrafted Lgr5-derived cells after transplantation.
The current in vitro organoid culture system is an important tool to understand the biology of liver stem cells. It should be emphasized that this model represents the bipotentiality of a single cell and can now allow interrogation of the biology of stemness, differentiation, and maturation. Furthermore, assuming that the engraftment pitfalls can be adequately addressed and the differentiation protocols optimized, these adult organ-derived cells may provide an important candidate for tissue engineering and regenerative therapies.
The appearance of Lgr5+ stem cells in the liver following injury is intriguing since this marker has shown to be expressed in stem cells of the gut, hair follicles, and other tissues. Based on the presented injury models, Lgr5+ cells may represent a dynamic stem cell compartment for hepatic repair as well. Several possible origins for these cells are outlined in Fig. 2, and there may be alternate scenarios that are not fully understood at this time. Whatever the source, the relative contribution of Lgr5+ progenitors to either cell compartment appears to be context-specific, depending on the mode and severity of hepatic injury. In addition, the exact mechanism by which Lgr5 may be regulating stemness remains a mystery.
The study also reports an important role of Wnt signaling in the maintenance and expansion of Lgr5+ stem cells. It should be noted that Lgr5 also modulates Wnt signaling, since binding of Rspo1 to Lgr5 induces interaction with and enhances internalization of Wnt coreceptors LRP6 and Frizzled. Since the precise sequence of events leading to activation of Lgr5+ cells in vivo is unknown, it would be of crucial significance to identify these signals and mechanisms. It is relevant to point out that canonical Wnt signaling promotes biliary epithelial cell proliferation and survival, similar to the Lgr5+ cells.[11, 12] Thus, it will be relevant to test if activation of Wnt signaling may be sufficient to reprogram all or a subset of biliary epithelial cells by inducing Lgr5 expression to initiate stemness.
The similarity of Lgr5 to Foxl1 as a marker for a putative liver stem cell population is worth emphasizing, since both have been shown to capable of self-renewal and bipotential differentiation. Shin et al. showed that the Foxl1+ population of progenitor cells was induced following a DDC diet, and appeared in the periportal region at the site of ductular reaction, suggesting that, like Lgr5+, these cells may arise from cells of biliary origin. Interestingly, Foxl1 appears to promote liver repair after bile duct ligation-induced liver injury through activation of the Wnt/β-catenin pathway, which stimulates proliferation of both hepatocytes and biliary epithelial cells. Although it appears that there is significant functional overlap in these two progenitor populations, the relationship between the Foxl1+ and the Lgr5+ populations remains an enigma.
Kari N. Nejak-Bowen, M.B.A., Ph.D.1
Satdarshan P.S. Monga, M.D.1,2
1Department of Pathology University of Pittsburgh, School of Medicine Pittsburgh, PA
2Department of Medicine University of Pittsburgh, School of Medicine Pittsburgh, PA